© Vattenfall AB

Decreased corrosion at high temperature?

Superheater corrosion testing in boilers Oslo, 16 FebruaryAnnika Stålenheim, Vattenfall Research and Development

© Vattenfall AB 22

Vattenfall in brief

• Europe’s fifth largest generator of electricity and the largest producer of heat

• Operations in Sweden, Finland, Denmark, Germany, Poland, the Netherlands, Belgium and the UK with a total of 7.4 million electricity customers and 5.6 million network customers

• 39,000 employees

• Vattenfall AB is wholly owned by the Swedish state

© Vattenfall AB 3

Background, higher steam temperatures

• Vattenfall plans for a dramatic increase of the biomass use Much biomass can be co-combusted in coal power plants

• There will also be a need for dedicated biomass plants• These plants should have high electrical and fuel efficiency to

ensure cost-effective biomass use• Vattenfall plans for at least 10 new bio-CHP plants

Increase the electrical efficiency– Today: approx. 30-31% net electrical efficiency, 540°C/140 bar

Potential: 36% net electrical efficiency, 600°C/190 bar– The boiler is the most critical component (corrosion at

superheaters and waterwalls)- This requires the right choice of materials + additives

© Vattenfall AB 4

Corrosion testing within WP 2.4.3

• Two corrosion probes (440°C and 490°C) with 11 different materials exposed near the superheaters in a waste fired boiler at MVB, Hamburg.

• Superheater corrosion probe used in Idbäcken, a BFB running on waste wood. Five different materials and a temperature gradient ranging from 450°C to 600°C.

• Waterwall test panels used in Idbäcken. Four different coatings were tested.

• Corrosion probes used before and after the ESP in Idbäcken. Two temperature gradient probes and one single temperature probe with seven different materials.

© Vattenfall AB 5

Müllverwertung Borsigstrasse (MVB)

• Located in Hamburg, Germany

• Waste fired grate boiler• 70% household waste• 57 MW heat and steam• Outgoing steam data:

380°C and 19 bar.

© Vattenfall AB 6

Two corrosion probes with 11 different materials exposed near the superheaters.

The metal temperatures of 440ºC and 490ºC correspond to steam temperatures of about 410ºC and 460ºC.

IACM

FTIR, deposit probes Corrosion probes

Corrosion probe testing at MVB, Hamburg

© Vattenfall AB 7

Chemical composition of the tested alloysAlloy Cr Fe Ni Other

15Mo3 0.1 Bal. 0.09 Al 0.007, C 0.17, Cu 0.12, Mn 0.67, Mo 0.29, P 0.006, S 0.004, Si 0.22

13CrMo44 0.85 Bal. 0.10 Al 0.01, C 0.12, Cu 0.12, Mn 0.46, Si 0.23, Mo 0.5, P 0.007, S 0.003

HCM12A 11.25 Bal. 0.5 B 0.005, Mn 0.45, Si 0.5, Mo 0.45, N 0.07, Nb 0.07, V 0.25, W 2.0

Super 304 18 Bal. 9 Cu 3.0, Mn 0.8, N 0.1 , Nb 0.45

317L cladding 18.5 Bal. 14.5 C 0.03, Mn 1,7, Mo 3.1, Si 0.4

Sanicro 28 26.58 Bal. 30.64 C 0.009, Cu 0.94, Mn 1.63, Mo 3.31, N 0.046, P 0.019, S 0.0005

Sanicro 41 (alloy 825) 19.88 36.1 38.25 Al 0.06, C 0.021, Cu 1.56, Mn 0.54, Mo 2.52, S 0.0005, Si 0.27, Ti 0.66

C-276 15.54 5.24 Bal. C 0.003, Co 1.20, Mn 0.50, Mo 15.70, P 0.007, S 0.003, V 0.2, W 3.50

C-22 21.47 3.76 Bal. C 0.004, Co 1.36, Mn 0.24, Mo 13.09, P 0.005, S 0.003, Si 0.027, V 0.12, W 3.18

C-2000 22.54 1.22 Bal. Al 0.24, C 0.007, Cu 1.53, Mn 0.23, Mo 15.65, P 0.004, Si 0.06

Alloy 625 cladding 21.4 1.0 64.5 Al 0.033, C 0.014, Mn 0.39, Mo 8.5, P 0.003, S 0.0021, Ti 0.21,

© Vattenfall AB 8

Material loss at two material temperatures

0

200

400

600

800

1000

1200

140015

Mo3

13C

rMo4

4

HC

M12

A

Supe

r 304

317L

cla

ddin

g

San

28

San

41

C-2

76

C-2

2

C-2

000

Allo

y625

cla

ddin

g

Cor

rosi

on p

er 1

000

h (µ

m)

Metal loss max 490C

Metal loss average 490C

Metal loss max 440C

Metal loss average 440C

Increasing Cr+Ni content

Ni-based alloys

© Vattenfall AB 9

Conclusions MVB

• The low alloyed steel 15Mo3, which is commonly used as a superheater steel in waste-fired boilers, performed well at a metal temperature 440°C in spite of its lack of protective alloying elements, but is unsuitable for use at higher temperatures.

• The alloys that performed best, especially at the higher metal temperature of 490°C, were, as expected, nickel base alloys containing chromium and molybdenum, such as alloys C-22, C-276, C-2000 and 625. These performed better at 490°C than 15Mo3 did at 440°C.

• The high strength austenitic stainless steel Super 304 performedexceedingly badly.

• Alloys C-22, C-276, C-2000 and Sanicro 41 performed better at 490°C than at 440°C.

Corrosion probe tests have been performed near the superheaters in a grate-fired boiler burning household waste. The temperature of the probes was controlled to 440°C and 490°C, simulating 410°C and 460°C steam and the tests lasted 1550 hours.

© Vattenfall AB 10

Idbäcken CHP (boiler 3)

Located in Nyköping, SwedenA BFB boiler producing 70 MWheat 35 MWelectricity

540°C/140 bar steam

Running on 100% waste woodPermanent installation of ChlorOut (sulphate additive), used for corrosion reduction

© Vattenfall AB 11

Fuel history of Idbäcken

0

100000

200000

300000

400000

500000

600000

2000

2001

2002

2003

2004

2005

2006

2007

2008

Waste wood MWhVirgin wood MWhOil MWhCoal MWhPower sold MWhHeat sold MWh

© Vattenfall AB 12

Corrosion testing at Idbäcken CHP

SH1

SH3

SH2

Fuel

ChlorOut

Airtpreheater

Air

Eco

Flue gascondensor

Elektrostaticfilter

Flue gas fan

• Corrosion probes

• Test panels

• Deposit analysis

• Flue gas analysis

© Vattenfall AB 13

Materials used

Material Cr Fe Ni Övriga 10CrMo910 2.25 Bal ≤ 0.50 C 0.12, Si ≤ 0.40, Mn 0.55, Mo 1.00,

Cu ≤ 0.30 Esshete 1250 15 Bal 9.5 C 0.1, Si 0.5, Mn 6.3, P≤ 0.035, S ≤

0.015, Mo 1.0, V 0.3, Nb 1.0, B 0.005 347H 18 Bal 11 C 0.06, Si ≤0.75, Mn ≤2.0, P≤0.045,

Nb ≥ 8*C; ≤ 1.0 AC66 27 Bal 32 C 0.06 Sanicro 63 21 3 Bal (63) Mo 8.5, Nb 3.4

© Vattenfall AB 14

Material loss

0

50

100

150

200

250

300

350

400

450

500

10CrM

o910

347H

E 1250

10CrM

o910

347H

E1250

AC6634

7HE 12

50San

6334

7HE 12

50AC6634

7HE12

50San

6334

7HE 12

50

Material and position on probe

Cor

rosi

on p

er 1

000

h (µ

m)

Internal corrosion

Metal loss

600°C450°C

© Vattenfall AB 15

Material loss vs temperature, Esshete 1250

Esshete 1250

0

50

100

150

200

250

300

500 510 520 530 540 550 560 570 580 590 600 610Average Metal Temperature °C

Cor

rosi

on in

µm

/100

0 h

Max. measuredAverage of 24 pointsPoly. (Max. measured)Poly. (Average of 24 points)

Data for 610°C from a previoustest at Idbäcken is included in the diagram. 610°C corresponds to a steam temperature of 580°C, fairly close to the goal of 600°C steam.

© Vattenfall AB 16

Conclusions

• Corrosion rate for 10CrMo910 was very high – 490µm/1000 hours => ca 3,9mm/7000h

• The austenitic steels TP 347H, Esshete 1250 and AC66 as well as the nickel base material Sanicro 63 had relatively low corrosion rates. AC66 is the material showing the best results.

• The austenitic steels and the nickel-base alloy showed lower corrosion rates in the region of 600°C metal temperature, than at 580°C or below, indicating that it is possible to reach high steam temperatures when burning 100% waste wood.

Corrosion probe testing, simulating superheater corrosion was performed in a boiler firing 100% waste wood and using ChlorOut (ammonium sulphate).

© Vattenfall AB 17

Superheater corrosion and temperature

• It has been shown that increasing the material temperature can actually result in lower corrosion when using corrosion resistant materials.

• There is a logical explanation for this. The chloride level in the deposits decreases as the temperature increases and it is believed that the stainless steels can more easily form a protective chromium oxide layer at higher temperatures.

• The results indicate that it is possible to reach high steam temperatures when burning 100% waste wood and other bio fuels. A combination of ChlorOut and corrosion resistant steels is believed to be useful in achieving this.

© Vattenfall AB

Thank you!